MAT correction issue - leaning out on hot starts
 

Hi guys,

I've been having trouble with really lean hot starts and noticed that my MAT correction table is just 0% for all the values. The car runs great once its moving a bit after a being restarted.

Is this table still used in the most recent updates? I'm using DIYPNP for a 1990 1.6.

Should I use brain's table as a baseline?

Attachment 184888

Unless MAT is either falsely or actually high, then MAT Corr. will not help. I don't think it's the right solution anyway. MAT Corr. should be used for MAT Corr.

High or low impedance injectors?

Yes. Additionally, move the sensor to the cold side of the IC if possible.

Pretty sure nobody uses low impedance injectors with MS. Nobody smart anyway.

This. Makes a huge difference and it cools off much quicker once you start moving.

derp post.

I will explain. When I was N/A, I had heat soak that gave lean hot, due to false reading of AIT sensor to values much higher than ever actual. In that case, adding MAT correction was helpful and valid.

Now that I am boosted, and the AIT is in the cold side of the FMIC, there is no heat soak, and the Intake Air temps are correct, low, close to ambient even on hot restarts. Under that condition, I still get lean hot restarts. If I added MAT correction, it would never take effect, as the MAT does not go high, even though AFR goes lean.

In other words, Lean hot restart, AIT is still 70*, same as original start, same as after things re-stablilize. Hence, adding fuel via MAT correction at, say 100* will have no effect, as that part of the curve is not used.

Same as you said before, with AIT sensor located in intercooler, there is no heat soak. That has been my experience.

So, I stand by the statement, that if AIT is not going high, either falsely, or accurately, then adding fuel at the higher end of MAT correction will have no benefit. If one does have heat soak, such that AIT is reporting high temps, then MAT correction will help, as there is a variable that relates to the problem.

DNM

i will explain:

the ideal gas law code of the MS is aggressive at best.

If you're reporting intake temps higher than 80°F, real or not, the MS is going to pull a lot of fuel and the only way to counter the effect is to tune out the ideal gas law algorithm by using the MAT corrections table to zero them out and then fine tune them yourself.

This was a huge issue when MS3 first came out as they change the implementation from MS2. It was such a sore point for so many installs, that the developers went back and reworked how the MAT corrections and ideal gas law code works on the newer MS3 firmwares.

I've also gone as far as implementing 2 ait sensors in my ms3; measuring temperatures in various locations in the engine bay, and even with the AIT sensor directly after the in IC, there's still PLENTY of heat there, enough to cause intake temps to soar without airflow--not falsely.

In your case, you can tune your startup settings (ASE and Warmup); which are based on CLT or AIT temp. But i can't imagine anyone having a lean start issue at 70°F; that just doesn't really happen.

Also no one should ever be running batch injection and wasted spark; that elminates almost all lean issues where you can idle smoothly at 16:1.

Scott. I appreciate the edit.

I am in the process of tuning EAE, which is quite the challenge. It tends to mess with my start, so perhaps my restart. If that is the case, as I get EAE dialed in, my issues will likely be fixed.

Re: Lean re-starts when AIT is 70*. I agree it does not make sense to me either. Hence my looking at other causes than MAT correction.

My MAT is flat from 60* to 90*, no fuel pull at 80*.

But, I don't want to hijack this thread, so I will bow out.

I'm using 550cc rx7 injectors.

I'm not sure how big of a difference it will make moving the sensor from it's current location. The IC pipe it's on doesn't get much cooler than the end tank, they are both downstream and within 6-8in or so of each other.

During a hot start - the car will go all the way to 19.0:1 and stall if i don't give it a little gas. Once it's moving/cooled slightly it will richen up and run fine.

you need to tune the MAT corrections to compensate.

Was this approach successful? I recently had problems with the engine going very lean on hot restarts. This is with one of the pre 1.3.1 builds (upgrading to latest 1.3.1 update in near future). We were using the Miata for drifting, so very high throttle for maybe 5 minutes with relatively low to road/air speeds to cool things down between runs. IAT temps were 150-160 on restart. I was running the "50%" MAT table to begin with, but then cut that in half again.

I noticed very little change in behavior after changing the MAT table. The car would run lean for a few minutes or until the driver drove around the pit area, at which point it would settle back down to normal range AFR's.

What we eventually noticed was a lot of pressure build up in the gas tank. If we cracked the gas cap to release the pressure, the lean start condition would clear up. This is with a NB2 VVT engine swapped into a 97 NA with no charcoal canister (vent line is connected directly to the OEM line which runs into the front subframe, previously used by the charcoal canister). I haven't confirmed, but believe that line isn't actually allowing pressure to release.

My plan is to go ahead and install a new fuel HP/flow fuel pump, new pressure regulator/gauge, etc and check that the vent line is able to actually vent.

This may not have any connection to the OP's problem, but thought it was worth sharing in case it "sparked" some ideas for the OP.

I have the same issue with hot-starts, or even just idling with a hot engine on my '99. Car goes very, very lean in idle after an Autocross run.
In my case it's NOT the MAT correction- the intake air temp reads close to ambient and I dialed the impact of MAT correction waaaay down anyway.

My latest theory is that the injectors heat-soak and their dead time increases a little as a result.
Since the '99 has a no-return system, there's no cool gas from the tank cooling the injectors. Not a big deal on the stock injectors, where 0.1ms or 0.2ms difference in dead-time don't have a huge impact on fuel delivered, but really important on my rather large (850cc) injectors.
Will eventually convert to a return system to confirm.

I have same issue.... Air Temp sensor is located on cold side of intercooler and reads ambient on hot re-start.

O2 readings go from the regular 13AFR at idle to somewhere near 16-17AFR...car stalls at idle for the first few minutes.

Injectors are DW1000 and dead time is very high (1.62ms @13.2v) to begin with.

I'm thinking injector heat soak on returnless system as well but i'm currently just running a richer idle to compensate.

how you logged a good idle vs. the hot idle to see if there's any difference?

any enrichments? same VE%? etc. etc.

My Megalog viewer won't open any logs from Tunerstudio...haven't logged. Something about X axis not defined...haven't looked into it yet... Haven't noticed any compensation setting differences in TunerStudio though...

In my case there's no fuel being pulled, rpm-range is the same, MAP is the same, pulse-width is the same. It's not the MS pulling fuel.
At identical settings less fuel is being delivered at high temperatures under the hood, or for some reason the car needs more fuel (doubtful).

Opening the hood actually helps- presumably because hot air extracts a little better with the hood open and the injectors can cool down.

Maybe I'll throw some ice cubes on my injectors and see if that helps:party:

Just checked and here's the result:
At 14.5V, which is where my car idles, my programmed injector dead-time is 1.21ms.
Pulse-width is 2.3ms to 2.35ms in normal idle.
Doesn't seem out of the ordinary to me. I usually idle around 12AFR, but going to 17 and even stalling when it get's nice and toasty under the hood.

does the purge solenoid allow the gas tank to vent on a returnless?

Tank is currently vented to air (purge deleted and hose points towards ground to vent tank)